Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “supervised training”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 181 records · Page 10

The Power of Priors: Improved Enrichment Safeguards

International safeguards currently rely on material accountancy to verify that declared nuclear material is present and unmodified. Although effective, material accountancy for large bulk facilities can be expensive to implement due to the high precision instrumentation required to meet regulatory targets. Process monitoring has long been considered to improve material accountancy. However, effective integration of process monitoring has been met with mixed results. Given the large successes in other domains, machine learning may present a solution for process monitoring integration. Past work has shown that unsupervised approaches struggle due to measurement error. Although not studied in depth for a safeguards context, supervised approaches often have poor generalization for unseen classes of data (e.g., unseen material loss patterns). This work shows that engineered datasets, when used for training, can improve the generalization of supervised approaches. Further, the underlying models needed to generate these datasets need only accurately model certain high importance features.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Literature mining of protein-residue associations with graph rules learned through distant supervision

Background: We propose a method for automatic extraction of protein-specific residue mentions from the biomedical literature. The method searches text for mentions of amino acids at specific sequence positions and attempts to correctly associate each mention with a protein also named in the text. The methods presented in this work will enable improved protein functional site extraction from articles, ultimately supporting protein function prediction. Our method made use of linguistic patterns for identifying the amino acid residue mentions in text. Further, we applied an automated graph-based method to learn syntactic patterns corresponding to protein-residue pairs mentioned in the text. We finally present an approach to automated construction of relevant training and test data using the distant supervision model. Results: The performance of the method was assessed by extracting protein-residue relations from a new automatically generated test set of sentences containing high confidence examples found using distant supervision. It achieved a F-measure of 0.84 on automatically created silver corpus and 0.79 on a manually annotated gold data set for this task, outperforming previous methods. Conclusions: The primary contributions of this work are to (1) demonstrate the effectiveness of distant supervision for automatic creation of training data for protein-residue relation extraction, substantially reducing the effort and time involved in manual annotation of a data set and (2) show that the graph-based relation extraction approach we used generalizes well to the problem of protein-residue association extraction. This work paves the way towards effective extraction of protein functional residues from the literature.

59 BASIC BIOLOGICAL SCIENCES↗

Event‐Based Training in Label‐Limited Regimes

Abstract The distribution of attributes assigned using data on independent sensors for a specific source, for example, magnitude, can be richly descriptive for final event characterization and associated uncertainty. Attribute distributions can also provide powerful context for event characterization in the absence of comprehensive annotation. This work develops a way to leverage distributional information across a set of sensors in the absence of comprehensive annotation as a domain‐informed regularization term applied during gradient‐based learning. The regularization term is the basis of event‐based training which I show can be a powerful semi‐supervised learning (SSL) approach. I first use a simple feed forward neural network and a toy data set to outline how data set structure interacts with the assumptions inherent to many semi‐supervised learning approaches. I then demonstrate the effectiveness of event‐based training using a deep convolutional neural network for seismic event classification in Utah, which increases SSL accuracy from 92% to 97% on event classification with a limited number of training labels.

Linville, Lisa M.↗

TCR-H: explainable machine learning prediction of T-cell receptor epitope binding on unseen datasets

Artificial-intelligence and machine-learning (AI/ML) approaches to predicting T-cell receptor (TCR)-epitope specificity achieve high performance metrics on test datasets which include sequences that are also part of the training set but fail to generalize to test sets consisting of epitopes and TCRs that are absent from the training set, i.e., are ‘unseen’ during training of the ML model. We present TCR-H, a supervised classification Support Vector Machines model using physicochemical features trained on the largest dataset available to date using only experimentally validated non-binders as negative datapoints. TCR-H exhibits an area under the curve of the receiver-operator characteristic (AUC of ROC) of 0.87 for epitope ‘hard splitting’ (i.e., on test sets with all epitopes unseen during ML training), 0.92 for TCR hard splitting and 0.89 for ‘strict splitting’ in which neither the epitopes nor the TCRs in the test set are seen in the training data. Furthermore, we employ the SHAP (Shapley additive explanations) eXplainable AI (XAI) method for post hoc interrogation to interpret the models trained with different hard splits, shedding light on the key physiochemical features driving model predictions. TCR-H thus represents a significant step towards general applicability and explainability of epitope:TCR specificity prediction.

60 APPLIED LIFE SCIENCES↗

Resimulation-based self-supervised learning for pretraining physics foundation models

Self-supervised learning (SSL) is at the core of training modern large machine learning models, providing a scheme for learning powerful representations that can be used in a variety of downstream tasks. However, SSL strategies must be adapted to the type of training data and downstream tasks required. We propose resimulation-based self-supervised representation learning (RS3L), a novel simulation-based SSL strategy that employs a method of resimulation to drive data augmentation for contrastive learning in the physical sciences, particularly, in fields that rely on stochastic simulators. By intervening in the middle of the simulation process and rerunning simulation components downstream of the intervention, we generate multiple realizations of an event, thus producing a set of augmentations covering all physics-driven variations available in the simulator. Using experiments from high-energy physics, we explore how this strategy may enable the development of a foundation model; we show how RS3L pretraining enables powerful performance in downstream tasks such as discrimination of a variety of objects and uncertainty mitigation. In addition to our results, we make the RS3L dataset publicly available for further studies on how to improve SSL strategies.

97 MATHEMATICS AND COMPUTING↗

In-process monitoring and prediction of droplet quality in droplet-on-demand liquid metal jetting additive manufacturing using machine learning

Abstract In droplet-on-demand liquid metal jetting (DoD-LMJ) additive manufacturing, complex physical interactions govern the droplet characteristics, such as size, velocity, and shape. These droplet characteristics, in turn, determine the functional quality of the printed parts. Hence, to ensure repeatable and reliable part quality it is necessary to monitor and control the droplet characteristics. Existing approaches for in-situ monitoring of droplet behavior in DoD-LMJ rely on high-speed imaging sensors. The resulting high volume of droplet images acquired is computationally demanding to analyze and hinders real-time control of the process. To overcome this challenge, the objective of this work is to use time series data acquired from an in-process millimeter-wave sensor for predicting the size, velocity, and shape characteristics of droplets in DoD-LMJ process. As opposed to high-speed imaging, this sensor produces data-efficient time series signatures that allows rapid, real-time process monitoring. We devise machine learning models that use the millimeter-wave sensor data to predict the droplet characteristics. Specifically, we developed multilayer perceptron-based non-linear autoregressive models to predict the size and velocity of droplets. Likewise, a supervised machine learning model was trained to classify the droplet shape using the frequency spectrum information contained in the millimeter-wave sensor signatures. High-speed imaging data served as ground truth for model training and validation. These models captured the droplet characteristics with a statistical fidelity exceeding 90%, and vastly outperformed conventional statistical modeling approaches. Thus, this work achieves a practically viable sensing approach for real-time quality monitoring of the DoD-LMJ process, in lieu of the existing data-intensive image-based techniques.

Gaikwad, Aniruddha (ORCID:0000000285642621)↗

L-PBF High-Throughput Data Pipeline Approach for Multi-modal Integration

Abstract Metal-based additive manufacturing requires active monitoring solutions for assessing part quality. Multiple sensors and data streams, however, generate large heterogeneous data sets that are impractical for manual assessment and characterization. In this work, an automated pipeline is developed that enables feature extraction from high-speed camera video and multi-modal data analysis. The framework removes the need for manual assessment through the utilization of deep learning techniques and training models in a weakly supervised paradigm. We demonstrate this pipeline’s capability over 700,000 high-speed camera frames. The pipeline successfully extracts melt pool and spatter geometries and links them to corresponding pyrometry, radiography, and processparameter information. 715 individual prints are examined to reveal melt pool areas that exceeds 0.07 mm 2 and pyrometry signal over a threshold (375 pyrometry units) were more likely to have defects. These automated processes enable massive throughput of characterization techniques.

36 MATERIALS SCIENCE↗

Extrusion parameter control optimization for DIW 3D printing using image analysis techniques

Material extrusion is a well-recognized facet of additive manufacturing that involves the fabrication of parts through the deposition of structural material from an extrusion head from a bulk supply. In the subdivision of Direct Ink Writing (DIW) additive manufacturing, challenges arise when the structural material is flowable, synchronous extrusion control and tool movement becomes critical for achieving high-quality parts with low defect populations. DIW techniques are most used in laboratory settings using expensive custom instruments and may require specialized 3D slicing software. Here, in this study, the fabrication of an inexpensive, consumer-friendly progressive cavity pump dispensing system is detailed, in which can create high-quality parts by executing G-code commands produced from a commercial slicing software. The precision and repeatability of the movement-synchronized material extrusion is demonstrated through a series of optimization schemes, entailing the alteration of various control parameters, which directly affect the extrusion properties demonstrated during a print. In situ diagnostics were implemented to evaluate the results of the established optimization experiment. Using a machine vision technique, images of the optimization prints are processed. Following this, a supervised machine learning model was trained to autonomously judge whether or not the extrusion parameters produced a passing or failing result. The machine learning scheme serves as a preliminary benchmark for future layer-by-layer evaluation of more complex DIW parts. The construction of the printer and development of in situ characterization capabilities demonstrates the ability for this printer to create high-fidelity DIW parts for a fraction of the price of other systems.

42 ENGINEERING↗

Supervised detection of anomalous light curves in massive astronomical catalogs

The development of synoptic sky surveys has led to a massive amount of data for which resources needed for analysis are beyond human capabilities. In order to process this information and to extract all possible knowledge, machine learning techniques become necessary. Here we present a new methodology to automatically discover unknown variable objects in large astronomical catalogs. With the aim of taking full advantage of all information we have about known objects, our method is based on a supervised algorithm. In particular, we train a random forest classifier using known variability classes of objects and obtain votes for each of the objects in the training set. We then model this voting distribution with a Bayesian network and obtain the joint voting distribution among the training objects. Consequently, an unknown object is considered as an outlier insofar it has a low joint probability. By leaving out one of the classes on the training set, we perform a validity test and show that when the random forest classifier attempts to classify unknown light curves (the class left out), it votes with an unusual distribution among the classes. This rare voting is detected by the Bayesian network and expressed as a low joint probability. Our method is suitable for exploring massive data sets given that the training process is performed offline. We tested our algorithm on 20 million light curves from the MACHO catalog and generated a list of anomalous candidates. After analysis, we divided the candidates into two main classes of outliers: artifacts and intrinsic outliers. Artifacts were principally due to air mass variation, seasonal variation, bad calibration, or instrumental errors and were consequently removed from our outlier list and added to the training set. After retraining, we selected about 4000 objects, which we passed to a post-analysis stage by performing a cross-match with all publicly available catalogs. Within these candidates we identified certain known but rare objects such as eclipsing Cepheids, blue variables, cataclysmic variables, and X-ray sources. For some outliers there was no additional information. Among them we identified three unknown variability types and a few individual outliers that will be followed up in order to perform a deeper analysis.

79 ASTRONOMY AND ASTROPHYSICS↗

Improving neutrino oscillation measurements through event classification

Precise neutrino energy reconstruction is essential for next-generation long-baseline oscillation experiments, yet current methods remain limited by large uncertainties in neutrino-nucleus interaction modeling. Even so, it is well established that different interaction channels produce systematically varying amounts of missing energy and therefore yield different reconstruction performance–information that standard calorimetric approaches do not exploit. We introduce a strategy that incorporates this structure by classifying events according to their underlying interaction type prior to energy reconstruction. Using supervised machine-learning techniques trained on labeled generator events, we leverage intrinsic kinematic differences among quasielastic scattering, meson-exchange current, resonance production, and deep-inelastic scattering processes. A cross-generator testing framework demonstrates that this classification approach is robust to microphysics mismodeling and, when applied to a simulated DUNE 𝜈 𝜇 disappearance analysis, yields improved accuracy and sensitivity at the 10%–20% level. These results highlight a practical path toward reducing reconstruction-driven systematics in future oscillation measurements.

Ellis, Sebastian A. R. [King's College, London (Un↗

Adaptive Learning for Concept Drift in Application Performance Modeling

Supervised learning is a promising approach for modeling the performance of applications running on large HPC systems. A key assumption in supervised learning is that the training and testing data are obtained under the same conditions. However, in production HPC systems these conditions might not hold because the conditions of the platform can change over time as a result of hardware degradation, hardware replacement, software upgrade, and configuration updates. These changes could alter the data distribution in a way that affects the accuracy of the predictive performance models and render them less useful; this phenomenon is referred to as concept drift. Ignoring concept drift can lead to suboptimal resource usage and decreased efficiency when those performance models are deployed for tuning and job scheduling in production systems. To address this issue, we propose a concept-drift-aware predictive modeling approach that comprises two components: (1) an online Bayesian changepoint detection method that can automatically identify the location of events that lead to concept drift in near-real time and (2) a moment-matching transformation inspired by transfer learning that converts the training data collected before the drift to be useful for retraining. We use application input/output performance data collected on Cori, a production supercomputing system at the National Energy Research Scientific Computing Center, to demonstrate the effectiveness of our approach. The results show that concept-drift-aware models obtain significant improvement in accuracy; the median absolute error of the best-performing Gaussian process regression improved by 58.8% when the proposed approaches were used.

adaptive learning↗

The role of chromatin state in intron retention: A case study in leveraging large scale deep learning models

Complex deep learning models trained on very large datasets have become key enabling tools for current research in natural language processing and computer vision. By providing pre-trained models that can be fine-tuned for specific applications, they enable researchers to create accurate models with minimal effort and computational resources. Large scale genomics deep learning models come in two flavors: the first are large language models of DNA sequences trained in a self-supervised fashion, similar to the corresponding natural language models; the second are supervised learning models that leverage large scale genomics datasets from ENCODE and other sources. We argue that these models are the equivalent of foundation models in natural language processing in their utility, as they encode within them chromatin state in its different aspects, providing useful representations that allow quick deployment of accurate models of gene regulation. We demonstrate this premise by leveraging the recently created Sei model to develop simple, interpretable models of intron retention, and demonstrate their advantage over models based on the DNA language model DNABERT-2. Our work also demonstrates the impact of chromatin state on the regulation of intron retention. Using representations learned by Sei, our model is able to discover the involvement of transcription factors and chromatin marks in regulating intron retention, providing better accuracy than a recently published custom model developed for this purpose.

Biochemistry & Molecular Biology↗

Sources of error in thematic classification of remotely sensed imagery

From a statistician's point of view, the input datasets are rarely examined to determine their underlying frequency distribution; it is just assumed that the data are normal enough, and that the deviations from normality are unimportant. It is not clear how deviations from a hypothetical multivariate normal might affect the power of the classification process, and there is ample evidence in the literature that, at a minimum, the spectral channels are correlated. From a practioner's point of view, in a supervised classification the number of training fields for developing a statistical description of a given class is usually arbitrary. It is unclear how small changes in the details of the training field selection process affect the quality of the derived thematic information. The start of an examination of this latter problem is discussed.

Star, Jeffrey L.↗

Classification

A supervised learning task involves constructing a mapping from input data (normally described by several features) to the appropriate outputs. Within supervised learning, one type of task is a classification learning task, in which each output is one or more classes to which the input belongs. In supervised learning, a set of training examples---examples with known output values---is used by a learning algorithm to generate a model. This model is intended to approximate the mapping between the inputs and outputs. This model can be used to generate predicted outputs for inputs that have not been seen before. For example, we may have data consisting of observations of sunspots. In a classification learning task, our goal may be to learn to classify sunspots into one of several types. Each example may correspond to one candidate sunspot with various measurements or just an image. A learning algorithm would use the supplied examples to generate a model that approximates the mapping between each supplied set of measurements and the type of sunspot. This model can then be used to classify previously unseen sunspots based on the candidate's measurements. This chapter discusses methods to perform machine learning, with examples involving astronomy.

Oza, Nikunj C.↗

Developing a Machine-Learning-Based Processing Framework for Twitter and Other Crowdsourced Data

Crowdsourced data streams such as Twitter and other social media are important sources of real-time and historical global information for Earth science applications. At the NASA Goddard Earth Sciences Data and Information Services Center (GES DISC), we have been exploring the Twitter data stream for its potential in augmenting the validation program of NASA's Global Precipitation Measurement (GPM) mission. To realize this potential, we need to increase the information density and enhance the quality of filtered precipitation tweets. We have implemented various components of a machine learning (ML)-based processing infrastructure for crowdsourced data that outputs, in this instance, useful and usable information derived from precipitation tweets. We have test enriched the Twitter stream with higher quality active tweets from those knowingly contributing to our effort and from existing crowdsourced programs (e.g., mPING, CoCoRaHS). We have experimented with various algorithms for processing tweets, including Naà ve Bayes, Convolutional Neural Network (CNN), Hierarchical Attention Network (HAN), and semi-supervised learning (with tri-training). Our current work focuses on (1) automated review of Earth science-related publications to determine relationships between discipline research needs and ML algorithms; (2) investigating Sequential Generative Adversarial Network (SeqGAN) for processing precipitation tweets for anomaly detection; and (3) managing crowdsourced data in a way that is compatible with existing NASA satellite data archives and using the data for ML applications. Key results include (1) network visualization of NLP-processed publications in various Earth science disciplines; (2) difference between GPM-linked, generated tweets and collected actual tweets that is small for GPM-determined light to moderate rain cases and high for GPM-determined heavy rain cases; and (3) identification of MongoDB for storing raw tweets and Zarr format for gridded tweets (compatible with GPM data). Our results have taken us a step closer to an operational ML-based tweet processing infrastructure and have already demonstrated that tweet-derived precipitation information is potentially useful for validation of Earth science satellite data.

Teng, William↗

A semi-supervised learning method to produce explainable radioisotope proportion estimates for NaI-based synthetic and measured gamma spectra

Quantifying the radioactive sources present in gamma spectra is an ever-present and growing national security mission and a time-consuming process for human analysts. While machine learning models exist that are trained to estimate radioisotope proportions in gamma spectra, few address the eventual need to provide explanatory outputs beyond the estimation task. In this work, we develop two machine learning models for a NaI detector measurements: one to perform the estimation task, and the other to characterize the first model’s ability to provide reasonable estimates. To ensure the first model exhibits a behavior that can be characterized by the second model, the first model is trained using a custom, semi-supervised loss function which constrains proportion estimates to be explainable in terms of a spectral reconstruction. The second auxiliary model is an out-of-distribution detection function (a type of meta-model) leveraging the proportion estimates of the first model to identify when a spectrum is sufficiently unique from the training domain and thus is out-of-scope for the model. In demonstrating the efficacy of this approach, we encourage the use of meta-models to better explain ML outputs used in radiation detection and increase trust.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Long-time integration of parametric evolution equations with physics-informed DeepONets

Ordinary and partial differential equations (ODEs/PDEs) play a paramount role in analyzing and simulating complex dynamic processes across all corners of science and engineering. In recent years machine learning tools are aspiring to introduce new effective ways of simulating such equations, however existing approaches are not able to reliably return stable and accurate predictions across long temporal horizons. We aim to address this challenge by introducing an effective framework for learning evolution operators that map random initial conditions to associated ODE/PDE solutions within a short time interval. Such operators can be parametrized by deep neural networks that are trained in an entirely self-supervised manner without requiring one to generate any paired input-output observations. Global long-time predictions across a range of initial conditions can be then obtained by iteratively evaluating the trained model using each prediction as the initial condition for the next evaluation step. Here, this introduces a new approach to temporal domain decomposition that is shown to be effective in performing accurate long-time simulations for a wide range of parametric ODE and PDE systems, from wave propagation, to reaction-diffusion dynamics and stiff chemical kinetics, introducing a new way of rapidly emulating non-equilibrium processes in science and engineering.

97 MATHEMATICS AND COMPUTING↗